Stanley Milgram's Six Degrees Of Separation Tested With Internet Email

Duncan Watts of Columbia and coworkers have repeated Stanley Migram's famous 6 degrees of separation study of 1967 by having people send email to someone they didn't know by sending it to someone who might know a person closer to the intended target. The internet has not decreased the number of degrees of separation between people.

When the researchers asked people why they did not participate, less than 1 percent replied that they could not think of anyone to send the e-mail message to, suggesting that most simply did not want to be bothered.

Thus, the researchers assumed that many more of the e-mail chains could have been completed. They calculated that half of them would have been finished in five steps or less if the first sender and the target lived in the same country, and seven steps otherwise.

Let me reword this: if only people were not so indifferent to requests they get from distant acquaintances we'd all be able to reach anyone in 6 steps. Well, news flash for the people who did this study: we are lucky that most people will not forward mail along in big chain mailings because spam junk mail is already a big enough problem as it is.

The original experiment by Milgram used physical mailing of packages.

The founding experiment in the science of social networks was performed in 1967. Stanley Milgram of Harvard University asked randomly selected people in Omaha, Nebraska, to send packages to a Boston stockbroker identified only by his name, occupation and rough location.

The first experiment required a far larger amount of effort on the part of those who were intermediaries.

Watts doesn't think the internet is generating a lot of new ties.

"In this experiment, the internet is simply the tool we use to transmit messages," Watts told New Scientist, in an email. "Compared with offline interactions like work, school, family, and community, I don't see email as being a particularly compelling medium for generating social ties."

I think that for some subpopulations the internet is creating many new bonds. There are people who are finding lovers and even future spouses on the internet. Also, some people are getting to know others virtually who wouldn't otherwise know each other. Plus, it is making it a lot easier to stay in contact with old friends or to find them again.

They'd need to use a much larger pool of recipients to discover what categories of people are better connected.

Melbourne travel agent Sally Stansmore, 32, was one of the 18 targets, which included a US university lecturer, an archival inspector in Estonia, an IT consultant in India, a Norwegian veterinarian and a Kalgoorlie policeman.

She said the 40 messages she received came mainly from people she knew well and "a few blasts from the past". She did not know where the email chains had originated. "Usually, I'd only know the last two (contacts)," she said.

But the connections that matter more are connections to powerful people.

It follows from this comment below that people who know a lot of people who do not know each other are better connected, all else equal.

Also, Watts said that people who were only remote acquaintances played an important role in the successful chains. That's because close friends tend to know the same people and have the same contacts, while more distant acquaintances are more apt to bring in new contacts unknown to the searcher.

One thing this study doesn't measure is whether or not people are more likely to be connected to others who share their own interests. Is the internet enabling relationship networks to be sorted more efficiently along the lines of common interests and beliefs?

Work acquaintances were more likely to pass along a message.

"Successful chains - in comparison with incomplete chains - disproportionately involved professional ties rather than friendship and familial relationships. Successful chains were also more likely to entail links that originated through work or higher education," the authors wrote.

Men passed messages more frequently to other men, and women to other women. This tendency to pass messages to a same-sex contacts was strengthened by about 3% if the target was the same gender as the sender and similarly weakened in the opposite case, the researchers found.

Another interesting result of the study was that highly social individuals who had large sets of acquaintances seem to be poor choices for sending messages to. It could be that they are too busy to send on messages. Or perhaps they are the center of an insular group.

Posted by Randall Parker at August 11, 2003 05:43 PM